Five million developers and a funding drought: building from Jakarta in the AI era

One of my friends is a software developer in Jakarta. Whenever people talk about where the next wave of builders will come from, they tend to name Silicon Valley, Bangalore or Shenzhen. Indonesia rarely comes up, which is strange given how many people there write code.

This piece looks at Indonesia's developer community: how big it has become, what happened when the investment money dried up, and what that means for the people writing the code. It isn't a pep talk. Big projects need money, and much of that money has gone away. But the same AI tools that let one person do the work of a small team change what a developer can build without it.

A developer boom, by the numbers

GitHub publishes country-level data through its Innovation Graph. In the first quarter of 2026 it counted about 5.27 million developer accounts in Indonesia, up from about 880,000 in early 2020. Counting countries rather than the EU as a bloc, that makes Indonesia the seventh-largest developer community on GitHub, just ahead of Germany. The figure you may have seen elsewhere, 4.37 million and eighth in the world, was Tech in Asia's report from late 2025, and GitHub's own data has grown past it since. These are accounts, not a census of professional programmers. Students, hobbyists and people who signed up once are all in there. Still, the trend is hard to argue with.

GitHub's annual Octoverse report says the same thing in its own words. In 2024 Indonesia had more than 3.5 million developers, growing 23% a year. In 2025 GitHub said Indonesia, India and Brazil had more than quadrupled their developer numbers since 2020. The same report notes that nearly 80% of new developers on GitHub use its AI coding assistant within their first week. The new generation of developers is learning to code with AI from day one.

Indonesians are also enthusiastic AI users. The Google, Temasek and Bain e-Conomy SEA 2025 report found that 80% of Indonesian users say they interact with AI tools every day, and that revenue for apps with AI features grew 127% between the first halves of 2024 and 2025, the fastest in Southeast Asia.

So there are millions of people writing code and millions of customers using AI. On paper, that is everything a tech industry needs. What's missing is the money.

The money went away

The same e-Conomy report includes a chart worth looking at closely. Private funding for Indonesia's digital economy peaked at $9.1 billion across 649 deals in 2021, then fell to $0.4 billion across 73 deals in 2024. The first half of 2025 brought about $0.1 billion across 20 deals. That's not a dip. It's a drop of more than 95% from the peak.

AI specifically hasn't filled the gap. Between mid-2024 and mid-2025, Indonesian AI companies raised $91 million, about 4% of AI investment across ASEAN. Google counts around 45 AI startups in Indonesia, against more than 495 in Singapore.

Some of the companies that did raise money left painful memories. GoTo, the country's biggest internet company, cut 1,300 jobs, 12% of its workforce, in late 2022. The agritech startup eFishery, once backed by SoftBank and Temasek, cut about 90% of its staff in early 2025 after an investigation found it may have inflated its revenue for years. A collapse like that makes investors more cautious about everyone else.

What a funding drought costs

It's worth being plain about this. Some things really do need money. A national-scale payments network, a hospital system or a large AI model needs servers, staff, lawyers and years of runway. Without investment, many of those projects in Indonesia will start late, start small, or not start at all. Being good at code doesn't change that.

The drought also changes where talent goes. When local startups stopped hiring aggressively, pay followed. The Monks Hill Ventures and Glints Startup Talent Trends Report 2024 found that Indonesian startup salaries fell about 3% on average, and 5% for junior staff, with engineering pay hit hardest, down 7%. CNBC Indonesia's write-up of the report puts a startup developer with one to three years of experience at roughly $600 to $800 a month.

At the same time, many skilled developers now work from home in Indonesia for companies based somewhere else. An LSE researcher who interviewed Indonesian Upwork freelancers found that one common path was becoming a permanent employee who works remotely for a particular company, often a foreign client they met on the platform. Kompas profiled one remote worker whose monthly pay from a foreign company is far higher than what he earned in an office job in Jakarta.

That's a good deal for the individual, and nobody should begrudge it. But it's also a kind of talent bleed. The developer stays home, but their best hours go into products owned and sold somewhere else. The local business down the street, which could have used that skill, never gets it.

One more honest data point: online freelancing isn't a gold mine either. The same LSE study found fewer than 10,000 searchable Indonesian Upwork profiles that had earned at least a dollar, and only 169 that had earned more than $100,000. Competing for global gigs against the whole world is hard.

The other side of the same coin

This is where the SoloPro angle comes in.

A few years ago, a product idea that needed a front end, a back end, payments, support and marketing needed a team, and a team needed funding. That's no longer true for a lot of useful software. GitHub's own assistant launched a free tier in late 2024 with 2,000 code completions and 50 chat messages a month.

That doesn't help anyone build a national payments network. It does help with the thousands of smaller, practical problems nobody is going to raise a Series A to solve.

And Indonesia has plenty of those problems. Bank Indonesia counts 65 million micro, small and medium enterprises, and 39.3 million merchants accepting QRIS, the national QR payment standard. Most of those merchants are small businesses, so they're already digital at the till. Yet one 2025 study, citing government data, found fewer than 10% of MSMEs use AI in their core operations. Businesses say they want help. In a survey commissioned by Amazon Web Services (vendor research, so read it with that in mind), 57% of Indonesian businesses named a lack of skilled people as the main thing holding back their AI use.

There's also a local-language edge. Most AI tools are built in English first. Sahabat-AI, an open model family from GoTo and Indosat built with AI Singapore, is tuned for Indonesian, Javanese, Sundanese, Batak Toba and Balinese, and anyone can download it. A developer who knows how a warung owner in Yogyakarta actually talks to customers knows something a team in San Francisco doesn't.

What a solo developer can actually build

None of this requires a pitch deck. A few realistic paths:

• Small AI-built products for one niche. Pick a group you understand, such as kos owners, tutoring centers, small clinics or batik sellers, and build one tool that does one job well. Charge a small monthly fee. A few hundred paying users is a real business for one person.
• AI setup as a service for local businesses. A shop that already accepts QRIS may still answer every customer question by hand. Set up and maintain things like a website chat bot, AI answers from a help page, booking without the back-and-forth or automatic invoice reminders. Then charge for keeping them running.
• Automation glue. A lot of small-business pain is copying information from one place to another. Tools like n8n and Make let one developer connect forms, spreadsheets, chat and payments in an afternoon. Developers who can also write code get further than the no-code crowd.
• Local-language versions of things that already exist. If a tool works well in English but badly in Bahasa Indonesia or Javanese, that gap is your product. Open models like Sahabat-AI make it cheaper to try.
• Keep the remote job, build on the side. There's nothing wrong with a steady contract. Many of the freelancers in the LSE study used platform work as a stepping stone, and one grew a presentation-design gig into a company with 55 staff. A side project that serves local customers builds something you own.

Keep costs low: free tiers, one cheap server, and payments through tools your customers already use. Don't take on debt to chase scale you don't need yet.

None of this replaces venture funding, and it would be dishonest to say it does. A solo developer won't build the next GoTo from a laptop. Solo businesses are small by design, and not everyone wants to run one. But it's a real way for a skilled developer to turn their skills into something they own, serving customers they can see.

What to take from this

Indonesia has done the hard part that money can't buy quickly: millions of people learned to code, and millions more are comfortable using AI. The money that was supposed to turn that talent into big companies mostly left, and pretending otherwise doesn't help anyone. Some big projects will wait. Some of the best developers will keep spending their best hours on other people's products.

But the tools that changed the math for big companies also changed it for individuals. A developer in Jakarta, Bandung or Surabaya no longer needs a funding round to ship something useful to the business down the street. Not having funding is a real constraint. It doesn't have to stop anyone.